Instructions to use Arisp/memorize-instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use Arisp/memorize-instruct with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3.5-2B-Base") model = PeftModel.from_pretrained(base_model, "Arisp/memorize-instruct") - Notebooks
- Google Colab
- Kaggle
Download sft_torch.json from Arisp/memorize-instruct: direct link, hf CLI and curl.
- Browser
- Download file 895 Bytes
-
https://huggingface.co/Arisp/memorize-instruct/resolve/main/sft_torch.json
- Command line
-
hf download hf://Arisp/memorize-instruct/sft_torch.json
-
curl -L -o sft_torch.json https://huggingface.co/Arisp/memorize-instruct/resolve/main/sft_torch.json
895 Bytes
| { | |
| "args": { | |
| "model": "Qwen/Qwen3.5-2B-Base", | |
| "data": "C:\\Users\\arisp\\Documents\\Research\\Memorize\\data\\sft", | |
| "out": "adapters/sft-run2-torch", | |
| "epochs": 1, | |
| "batch_size": 8, | |
| "max_batch_tokens": 4096, | |
| "lr": 1e-05, | |
| "clip": 20.0, | |
| "warmup": 20, | |
| "rank": 8, | |
| "scale": 20.0, | |
| "max_seq": 512, | |
| "report_every": 25, | |
| "eval_every": 100, | |
| "no_probes": false, | |
| "gpu_mem_frac": 0.9, | |
| "seed": 0 | |
| }, | |
| "history": [ | |
| { | |
| "step": 0, | |
| "valid_loss": 1.3611059830069272 | |
| }, | |
| { | |
| "step": 100, | |
| "valid_loss": 1.2591907709679355 | |
| }, | |
| { | |
| "step": 200, | |
| "valid_loss": 1.2521158879783705 | |
| }, | |
| { | |
| "step": 300, | |
| "valid_loss": 1.2502558373648405 | |
| }, | |
| { | |
| "step": 300, | |
| "valid_loss": 1.2502558373648405, | |
| "probes_passed": 18 | |
| } | |
| ] | |
| } |